CODEWITHSUNDEEP
rdataframe
french_fries
french_fries

Reputation: 1

Specific dataset subseting

I have a dataset:

x   y
j   5
k   2
b   9
b   1
k   1
g   8
h   2
k   4

I want to extract rows with value k in column x and row above it, so desired output is:

x   y
j   5
k   2
b   1
k   1
h   2
k   4

How could I do that? I know how to extract rows with value k in column x:

df[df$x=="k",]

, but how to extract also a row above it?

Upvotes: 0

Views: 39

Answers (3)

akrun
akrun

Reputation: 886938

One option with lead

library(dplyr)
df %>%  
    filter((x == 'k')|lead(x == 'k'))
#  x y
#1 j 5
#2 k 2
#3 b 1
#4 k 1
#5 h 2
#6 k 4

Or with slice

df %>%
     slice(which(x == 'k') + rep(c(-1, 0), sum(x == 'k')))

In base R

df[which(df$x == 'k') + rep(c(-1, 0), sum(df$x == 'k')),]

Upvotes: 1

Mohanasundaram
Mohanasundaram

Reputation: 2949

Extract the rownames and subset the data frame with the row names an row numbers as below

> df[sort(c(as.numeric(rownames(df[df$x=="k",]))-1, as.numeric(rownames(df[df$x=="k",])))),]
  x y
1 j 5
2 k 2
4 b 1
5 k 1
7 h 2
8 k 4

Upvotes: 0

ThomasIsCoding
ThomasIsCoding

Reputation: 101024

You can try the code below using which+ sort

indk <- which(df$x=="k")
df[sort(c(indk-1,indk)),]

which gives

  x y
1 j 5
2 k 2
4 b 1
5 k 1
7 h 2
8 k 4

Data

> dput(df)
structure(list(x = c("j", "k", "b", "b", "k", "g", "h", "k"), 
    y = c(5L, 2L, 9L, 1L, 1L, 8L, 2L, 4L)), class = "data.frame", row.names = c(NA,
-8L))

Upvotes: 2

Related Questions